Skip to main content.
July 22nd, 2006

Ryle on Knowing How and Knowing That

Ryle offers a regress argument for the impossibility of reducing knowing how to knowing that. Jason Stanley and Timothy Williamson in Knowing How (PDF) suggest a way to block the regress. I think Ryle anticipated their reply, and has something interesting to say about it. I’m not sure whether Ryle’s response works, but it is I think a response. (What I’m going to say is similar to what Alva Noe says in his Against Intellectualism (PDF), but I hope different enough to be worth saying.)

Here is Ryle’s original formulation of the regress argument.

The crucial objection to the intellectualist legend is this. The consideration of propositions is itself an operation the execution of which can be more or less intelligent, less or more stupid. But if, for any operation to be intelligently executed, a prior theoretical operation had first to be performed and performed intelligently, it would be a logical impossibility for anyone ever to break into the circle.

Jason and Tim summarise Ryle’s position as follows.

Premise 1: If one Fs, one employs knowledge-how to F.
Premise 2: If one employs knowledge that p, one contemplates the proposition that p.

They then interpret Ryle as arguing that these premises plus an assumption that knowing how to do something is knowledge that as leading to a regress, where the agent has to contemplate infinitely many propositions to do anything.

Their response is two-fold. First, they argue that premise 1 is only plausible if we restrict attention to actions done intentionally. “If Hannah wins a fair lottery, she still does not know how to win the lottery, since it was by sheer chance that she did so.” They then argue that premise 2 is not plausible if we read it as saying that one has to intentionally contemplate the proposition that p. This way the regress is blocked.

They are clearly right that premise 1 is too strong as it stands. And there is little textual evidence that Ryle intended anything that strong. But the particular weakening that Jason and Tim suggest is not the only plausible weakening. Indeed, it is not obviously the weakening that Ryle has in mind.

To see what other options, let us make up a new word. Say that an action is done agentively iff it is evaluable using, as Ryle puts it on page 25, concepts from “that family of concepts ordinarily summarised ‘intelligence’”. These include ‘clever’, ‘sensible’, ‘prudent’, ‘witty’, ‘judicious’, ‘silly’, ‘careless’, ‘rash’, ‘uninventive’, ‘slow’ and many more. (You might suspect that an action is agentive iff it is intentional, but I doubt it. At times only a dull person would pause to act intentionally rather than habitually.) Now I think you can make a case for each of the following premises.

Premise 1a: If one Fs agentively, one employs knowledge-how to F.
Premise 2a: If one employs knowledge that p, one contemplates the proposition that p, and this act of contemplation is agentive.

There is some textual evidence that Ryle intended Premise 2a. Indeed, the following passage not only supports that interpretation but supports the premise.

But what makes him consider the one maxim which is appropriate rather than any of the thousands that are not? Why does the hero not find himself calling to mind a cooking-recipe, or a rule of Formal Logic? Perhaps he does, but then his intellectual process is silly and not sensible. Intelligently reflecting how to act is, among other things, considering what is pertinent and disregarding what is inappropriate. (31)

Is Premise 1a true? I don’t think the examples that Jason and Tim raise tell against it. But while it seems plausible, it is far from clear how to argue for it. That’s why I don’t want to endorse Ryle’s regress argument.

More generally, there is a worry to Jason and Tim’s strategy here. Premise 1 is too strong as it stands, so it must be weakened. For Ryle’s purposes, we have to strengthen Premise 2 in a way so that together with Premise 1 (as modified) it generates a regress. Now Jason and Tim show quite convincingly, I think, that weakening Premise 1 so it only applies to intentional actions will be no good. But there may be other weakenings that work. I’ve suggested one, and I don’t know of any argument that there are no others that could work. So the spectre of the Rylean regress still hangs over any attempt to rehabilitate intellectualism.

Posted by Brian Weatherson in Uncategorized

10 Comments »

This entry was posted on Saturday, July 22nd, 2006 at 3:31 pm and is filed under Uncategorized. You can follow any responses to this entry through the comments RSS 2.0 feed. You can skip to the end and leave a response. Pinging is currently not allowed.

10 Responses to “Ryle on Knowing How and Knowing That”

  1. jasoncs says:

    Brian,

    I need some help seeing why premise 2a is any more plausible than premise 2. To go back to a Ginet type counterexample, I employ my knowledge that the door has a round handle when I open it. But in no sense, agentive or not, do I contemplate the proposition that the door has a round handle when I employ my knowledge that the door has a round handle in opening it.

    It’s not like I can felicitously say of someone who exercises their knowledge that the door has a round handle in opening the door that they sensibly (or intelligently) contemplated the proposition that the door has a round handle.

  2. Brian Weatherson says:

    Premise 2a isn’t meant to be more plausible than premise 2 – in fact it is logically stronger.

    But I didn’t think you were making this kind of objection in the first place. In the paper you don’t say that the problem with Ryle is Ginet’s example. I thought you agreed that Ryle could say something here, but what he said would get him in trouble with premise 1.

    In any case, let me say something directly about these cases. I think Ryle was wrong to use the word ‘contemplate’ here. But I do think that in order to use the knowledge that the door has a round handle, the agent has to form a representation that the door’s handle is round, and this representation has to be either conscious or easily accessible to consciousness. Now I wouldn’t call that contemplation, but I do think there’s some kind of relation the agent has to stand in to the proposition, and we can ask whether it was sensible or not for the agent to come to stand in that relation.

  3. jasoncs says:

    Brian,

    What we argue is that Ryle is not entitled to both Premise 1 and Premise 2 of his regress argument. Premise 1 is only plausible if one suitably restricts the range of the variable ‘F’ (most plausibly, to intentional actions). What Ginet’s example shows is that Premise 2 is only plausible if ‘contemplates a proposition’ is taken as a kind of action that is not intentional (as we say, it is only plausible if we think of ‘contemplating a proposition’ like ‘digesting food’ (p. 416)). But then you can’t place ‘contemplating a proposition’ in for ‘F’ in premise 1.

    So we took it that Ginet had already refuted premise 2. We contemplate a suggested riposte on behalf of Ryle, which is to construe ‘contemplation of a proposition’ as an action in a deflationary sense. But then we conclude that it the riposte is ineffectual, because, while it may save premise 2, it doesn’t allow the regress to go through (for the one can’t substitute ‘contemplation of a proposition’ in for ‘F’ in premise 1).

    As I said above, I don’t see how your suggestion helps Ryle here, since it doesn’t rescue premise 2. At most, you’re saying that, instead of ‘intentional action’, there is a slightly different restriction on the range of ‘F’ on the true reading of premise 1. But unless the contemplation of a proposition counts as one of these kinds of actions, you’re not helping Ryle.

    I’m not sure whether I understand your notion of an agentive action fully. But I don’t see how contemplating a proposition could count as one, unless digesting food also does (and ‘digesting food’ is clearly not an appropriate substitute for ‘F’ in premise 1). But maybe you could fill it out more…

  4. jasoncs says:

    Brian,

    You write about the Ginet case:

    “I do think that in order to use the knowledge that the door has a round handle, the agent has to form a representation that the door’s handle is round, and this representation has to be either conscious or easily accessible to consciousness. Now I wouldn’t call that contemplation, but I do think there’s some kind of relation the agent has to stand in to the proposition, and we can ask whether it was sensible or not for the agent to come to stand in that relation.”

    It seems to me that what is sensible or not is the agent’s decision to open the door. Once an agent chooses to open a door, she automatically employs all sorts of propositional knowledge in carrying out her decision. Similarly, what is sensible or not is the agent’s decision to eat dinner; not the subsequent digesting of the food.

  5. Brian Weatherson says:

    I’m not sure Ryle would agree that it is a fair assumption that the agent ‘automatically’ employs all sorts of propositional knowledge. For one thing, there is a possible agent that does not do this, but instead when she decides to open the door, starts thinking about Paris, rather than about, say, the shape of the door handle.

    Perhaps more importantly, I think Ryle thinks the difference between know how and know that just is that know how is constituted by the disposition to bring to mind the right kinds of things. Ryle doesn’t think that a mind that consists of nothing but facts would automatically connect any given fact with any given situation. If you think that the agent is set up so that she automatically connects some tasks with some stores of information, then I think you agree with Ryle that there is a distinctive kind of know how.

    On the 7:49 post, I think there are all kinds of ways to restrict the range of ‘F’. And I think the restriction Ryle has in mind is to actions that can be the subject of evaluative attitudes. And I think he’s right that, although most of us do this kind of thing well out of habit, it is the kind of thing that can in principle be done badly, and can in principle be criticised. So in that respect it really isn’t like digesting food. Or do you think that ‘contemplating a proposition’ is like ‘digesting food’ in every way.

  6. jasoncs says:

    Brian,

    Believe me, digesting food is certainly something one can do well or badly as well. So it cannot be that the criterion of intelligent action is that it can be done well or badly.

    There are many cases in which I employ my knowledge that in such a way that there is no distinct action of contemplating a proposition that is under my intelligent control (which is after all what is required to produce a vicious regress). Indeed, I’m finding it very hard to imagine what the picture of the mental here is supposed to be. Your idea is that it is conceptually possible to employ my knowledge that the door handle is round in seeking to open the door, yet fail to represent, even tacitly, the door handle? If so, either I never employed the knowledge in the first place, or the possibility you describe just shows that we regularly employ our propositional knowledge without representing information in your sense of “represent”.

  7. Brian Weatherson says:

    I agree that digesting food can be done well or badly. But I don’t think that any of the predicates on page 25 can apply to the process of digestion itself, as opposed to the decision to put food in one’s mouth and swallow. On the other hand, calling to mind a list of state capitals as a means to opening a door is rather silly. Even if it is unintentional, or not under direct control, or for that matter not under indirect control, it is still a sign of a confused mind, and thus epistemically criticisable.

  8. Jason_Streitfeld says:

    I don’t see any textual evidence that Ryle adopts either P1 or P2. On the one hand, he defines knowing-how in terms of intelligent behavior; on the other hand, he does not claim that every employment of knowing-that entails an act of contemplation. Rather, examples of knowing-that can just entail an acknowledgment or statement of a proposition (as when one recites the rules of chess). So a more accurate representation of his regress argument might look like this:

    P1: One Fs intelligently iff one employs knowledge how to F.
    P2: If one employs knowledge-that, one states or acknowledges a proposition.

    He then observes that, if knowing-how is reducible to knowing-that, then

    R: If one employs knowledge how to F, one employs knowledge that r is a rule for F-ing.

    It follows that, if knowing-how is reducible to knowing-that, one cannot act intelligently without stating or acknowledging a rule for that behavior. The problem is that propositions can be stated or acknowledged intelligently or unintelligently, appropriately or inappropriately. The rule must be intelligently acknowledged, which increases the number of intelligent acts by one. This second intelligent act would also have to be guided by another one, ad infinitum.

  9. Jason_Streitfeld says:

    I can see two possible responses to Ryle’s argument (as I’ve framed it):

    1. Reject the supposition that knowing-that entails the statement or acknowledgment of a proposition.

    2. Claim that any intelligent act just is a statement or acknowledgment of a rule for that behavior. So, a musical improvisation, for example, would count as an acknowledgment of a rule (or rules) for that performance.

  10. Jason_Streitfeld says:

    I have to correct my initial post here. I don’t think Ryle adopts the P1 I proposed. Rather, it should be:

    P1: If one Fs intelligently, one employs knowledge how to F.

    Ryle does not suppose that all employments of knowledge how to F are intelligent acts of F-ing. Rather, knowing how to F can consist in understanding a rule for F-ing, and this can be exhibited in any number of ways, such as in instructing others in the successful pursuit of F-ing.

    This correction doesn’t change the force or nature of Ryle’s regress argument, but it does protect Ryle against well-known objections.

Leave a Reply

You must be logged in to post a comment.